With a staff of just two people, Silver Bulletin is now the #3 Substack newsletter in Substack’s largest category, U.S. Politics. We even briefly hit #2 during the election peak. That ranking is generous to us in various ways — it happens to use metrics that favor us1. Also, it’s likely to fall. (Unlike most Substacks, our traffic is highly seasonal/cyclical.) Still, it’s something we’re proud of. Even with some inevitable churn … let’s be honest, probably a lot of churn … it gives us a great base to grow from as we plan for the future and consider expansion opportunities. This year went way better than I expected, and I sincerely appreciate your support.
Although we’re best known for our models — the election model and the NCAA Tournament model and more sports models to return soon — about half of our paid subscriptions and two-thirds of overall subscriptions are generated by our articles, a.k.a. newsletters, a.k.a. blog posts. So, I wanted to share a few words about those. I basically started out digital life as a blogger2 — FiveThirtyEight.com spun out of pseudonymous blog posts I wrote at Daily Kos3 — so it’s a natural format for me.
What’s the difference between an article, a newsletter, and a blog post? Admittedly, the distinctions are subtle. An “article” implies a higher level of refinement, ideally with a crisp, newsworthy takeaway. A “newsletter” suggests something almost diary-like that is updated regularly. A “blog” post, in my view, is mostly a stylistic characterization. Blogs are defined by their spikiness. They are sharp-elbowed — kiki, not bouba — sometimes straddling the boundary between analysis and opinion and presenting a strong authorial point of view.
My writing has always drawn something from each of these buckets. And however you want to characterize it, it’s always drawn lots of attention — positive and negative. Generally speaking, I think people operate from too much of a scarcity mindset and are too afraid of criticism from strangers.4 At least having a newsletter somewhat levels the playing field: the haters will go on hating, but the lovers can become subscribers.
So, I wanted to share a few tips about my philosophy and take you behind the scenes of the Silver Bulletin writing process. These are hardly trade secrets. But at this point, I’ve been through a lot of phases of writing on the Internet, including having worked in traditional newsrooms like the New York Times. People come to Substack from a variety of places — migrating from newsrooms, transitioning from social media, expanding out from academia, or graduating from a personal email list. While I’m a big advocate for the newsletter business — in fact, I now have a modest-sized equity stake in Substack — I encounter some Substacks that I think are punching below their weight relative to the quality of their content and others that are astonishingly successful. While I often recommend the newsletter approach to media friends who are thinking about their next steps, it isn’t for everyone.
So, which habits make for a successful newsletter? Here’s my advice, organized into seven broad categories.
1. Always. Be. Blogging.
I have to tell you that producing Silver Bulletin is really hard work. It’s often creatively fulfilling and enjoyable. This year, it was financially rewarding. Professionally, I’m the happiest I’ve been in years. But it’s a lot of work.
Maybe you’re a professional journalist or writer, and you’re thinking of starting a Substack as a gateway to a leisurely lifestyle. Write when you feel like it: maybe you’re expecting to turn out a real banger once every month or two, and then there will be some filler content that keeps readers up to date on your other projects. Sporadically, you’ll paywall some of this.
This approach is … perfectly fine! If you have a large, established audience and your bangers bang hard enough, you may even be able to earn a decent living from it.
But I’m not writing this guide for people who want to go into semi-retirement. Instead, it’s for people who want their newsletter to become a thing: a real business that’s taking at least some steps to maximize (and monetize) its audience.
If we’re being honest, that requires becoming a little bit obsessed. The film Glengarry Glen Ross includes a legendary scene in which Alec Baldwin admonishes his team of sad-sack real estate salesmen about the ABCs: Always Be Closing. No matter what your tactics are, the bottom line is that you need customers who will sign onto the bottom line.
My version of this is ABB: Always Be Blogging. The goal is to produce blog posts — excuse me, newsletters — with some regularity. Substack produces relatively linear rewards for the effort you put in.
If you asked me to chart out how I spent my time in the way that, for instance, the American Time Use Survey does, you’d find that I’m not actually “working” all that often. Sure, 40 or 50 hours a week. But I take most nights off from “work”. I travel a lot, both for work and for fun. I play a lot of poker, a wonderful but time-intensive habit. And I have my share of weekday leisurely lunches. I do tend to put in some working hours on weekend mornings and afternoons. Still, it’s not a crazy schedule.
But in another sense, I’m working all the freakin’ time. A casual morning jog? I’m probably thinking about the next few newsletter posts or listening to a podcast that might provide the inspiration for one. Poker nights? Well, there’s a lot of downtime in between hands. Unless the game is especially high stakes, I’m probably thinking about the newsletter, too. Even when I’m sleeping, I’m kind of working on the newsletter. Like a lot of writers, I often find that ideas crystalize overnight and are there for me when I wake up.
As a writer, it's important to give yourself space — writing time and thinking time. Time to explore the world, have conversations and generate ideas. If you’re the sort of person who looks at an empty calendar and is slightly terrified by it, this lifestyle is not for you.
Because you’ll have to immerse yourself in the process for long parts of the day. The best newsletters are more than the sum of their posts, offering a worldview and way of thinking.
Basically, the way that I look at blogging is that you’ll want to have a well-stocked pantry: lots of ingredients ready to go so that you can whip something up on short notice for a weary traveler. Something that might be pretty damned good, in fact. In the Silver Bulletin kitchen, there are always foodstuffs in various stages of preparation: half-written posts, ideas for which I’ve collected the data or made a few charts, and various bullet-pointed lists.
If you look at the top-performing politics newsletters, nearly all are turning around several articles a week, and some publish almost every day. You have to go down to Andrew Sullivan’s Weekly Dish at #15 to find an exception, and Andrew (whom I’ve gotten to know) is a unique case because he built up a massive audience through daily blogging when his newsletter was then called the Daily Dish.
While I’m trying to avoid being overly proscriptive, I strongly suggest there is an inflection point for a successful Substacker at publishing at least two items per week. Max Read, in an excellent post about his Substack process, came to the same conclusion. I also have some anecdata to back this up. When I was in the midst of finishing my book, I had some months where I was down to one post a week instead of the two or three I usually aim for. Those months were notably slower for subscription growth — basically just treading water to fight against the churn.
What’s the magic of 2+ items per week? Well, newsletters typically have a shelf life of two or three days. You write something. Some people read it right away. But others bookmark it for when they have a moment free. And for a post to travel beyond your email list, it usually takes another day or two. With two to three posts a week, each producing a traffic spike for two or three days, you’ll always have something that feels fresh.
And you’ll certainly want to publish at least one item a week. Let’s face it: on some level, successful newsletters depend on readers forming habitual relationships. It’s hard for things you do less than once a week to be habit-forming. The reason I suggest aiming for 2+ stories per week is that not every post is going to work for everyone. In particular, if you have both paid and free posts — and by the way, unless you don’t care about the money at all, I suggest that the ratio of paid-to-free posts be somewhere in the vicinity of 50:50 — writing enough so that your free subscribers get something worthwhile from you every week is valuable.5
2. Stretch singles into doubles
Here’s where Substackers who have worked in newsrooms have an edge. They know how to “punch up” a story, to make it sizzle. This does not require any sort of trickery. On the contrary, clickbait headlines may produce clicks — but you don’t want clicks; you want subscriptions. Misleading headlines may even produce unsubscribes instead.
But you need to sell your work. As the media landscape becomes more fragmented, there’s a premium on being entrepreneurial. I call this category “stretch singles into doubles” because it comes after you’ve already done the hard work of getting on base through the Always Be Blogging process. And there are four mechanisms to get those stretch doubles:
The headline
The lede — that is, the opening to your story
The preview image
And with how you promote it on social media
I’d recommend reviewing your posts once a month and looking at which overperformed or underperformed relative to your expectations. In your evaluation, consider a variety of indicators: total subscriptions generated, paid subscriptions generated, likes, open rates, views, and shares. Basically, everything except the number of comments, which isn’t well-correlated with the other metrics and sometimes just reflects that you’ve written on a hot-button topic or even pissed your readership off.
When I review my posts, I find the difference between more and less successful ones is often whether I put effort into the four categories I described a moment ago. This post on the changing nature of inflation never quite got the traction I was expecting, for instance — and I think that’s because it had a slightly weird headline and came in a phase when I was experimenting with AI-generated banner images that people seem to hate. By contrast, this post on pollster “herding” took off, in part because it has a compelling headline and preview image.
I’ve listed these categories in rough order of importance. Headlines are crucial. Be careful with this — but sometimes you’ll come up with such a good headline that the article almost writes itself. Far more people see the headline than read the article. And headlines dramatically affect their expectations if they do read on. Don’t forget that for an email newsletter, many people see the headline as a subject heading in their inbox. So short and simple often works best — e.g., “Joe Biden Should Drop Out” — although there’s room for creativity as in the “sheep farm” example.
The baseball fans among you might ask: why focus on this? Isn’t it more important to get on base? Well, yes, it is — the marginal value of reaching first base rather than making an out is considerably higher than getting to second.
But stretching singles into doubles is something you can do with relatively marginal effort. If you spend a whole day working on a post but then slap on a bad headline in a few seconds, you’re undermining all that effort you put in. Plus, although every writer has slumps, I’d try to avoid publishing stories that just barely meet the threshold of being worth the reader’s time. You’ll want everything to be solid contact.
3. Home runs come from timely, differentiated content
To extend the baseball metaphor, home runs are more predictable than any other type of hit. They come when a pitcher leaves a juicy pitch somewhere out over the middle of the plate. And the batter takes a big rip and connects.
The equivalent in the newsletter business is when there’s a topic that you have some unique expertise or insight on — and you’re at bat when the news comes in and prepared to take a big swing. The most popular post in Silver Bulletin history, outside of the model landing pages, is this one on Ann Selzer’s Iowa poll that — dubiously and wrongly! — had Kamala Harris leading in Iowa. That post came out relatively late on a Saturday evening, usually one of the worst times for newsletter traffic. But it got almost 350,000 views that night and an equal number the next day. I canceled my evening plans to turn it around as quickly as I could. I’m glad I did because it would have been an expensive night out, considering the opportunity cost.
I almost omitted the “timely” qualifier and went with “home runs come from differentiated content.” Isn’t timeliness just one of many ways to differentiate your writing? Yes, but it’s often the most important way. If you’ve worked in a newsroom, you’ll know that there’s always a trade-off between speed and depth. Being first to a story, or better yet, having the first authoritative take on the story, counts for a lot. It becomes the article that everyone else links to. It can even frame the entire public discussion about the topic.
It isn’t necessary to be a home run hitter to have a successful Substack. Some writers like Matt Yglesias and Heather Cox Richardson are the Ichiro Suzukis of the blogging ecosystem, with a high batting average. Still, if your value proposition is that you offer unique expertise on a particular subject, you’ll want to serve your readers when news comes up in a category that everyone on the Internet is talking about, and you have something distinctive to say about it. Even if it ruins your evening plans.
4. Measure twice and cut once
This is the most dangerous tip because it’s more about how one approaches writing than their newsletter specifically — and every writer is different. But I cringe when I see advice like: “Just start typing and figure it out on the page.”
Don’t do this. Most of the writing process is figuring out what you want to say. And if you haven’t figured it out, typing doesn’t help. Worse, you may become attached to your inadequate first draft. It often takes longer to improve bad copy than to start with a blank page.
So be prepared. Think about the post as you’re going through your morning routine. Then, you may want to draw up some sort of outline — personally, I don’t rigorously follow detailed outlines, but they remind me of the key points I want to hit, so a series of bullet points may suffice.
You’ll get better at estimating how long it takes you to write any post relative to the subject matter — if you’re not writing on the topic regularly, it will require more effort — and the depth and complexity of the post.
5. Take a nose-to-tail approach to the newsletter
In the culinary world, “nose-to-tail” is the philosophy that if you’re going to cook meat, you want to use the whole animal to avoid wastage (and the cruelty of killing any more living beings that you have to). Before, I spoke about the importance of having a well-stocked kitchen of ideas for your newsletter. This is the corollary and counterpart to that. Ideally, you don’t want to spend a bunch of time thinking about a subject and then not getting a newsletter or two out of it.
It isn’t easy. If, after two or three attempts, you find you’re just not getting traction on something — not turning your ideas into publishable copy — it’s probably time to abandon it. Often, it’s your subconscious telling you that your thesis isn’t clear or that the timing of the story isn’t right
Sometimes, the stories that stall out in the draft phase can have residual value as scrap material. If I have an idea about sports betting that isn’t quite working, for instance, but then there’s another betting scandal, I’ll probably be able to get something pretty good out of that work. Subscriber Q&As can also be good places for working with your scraps.
Although it might seem wasteful, I’d generally advise putting aside a story if something fresher hits your radar and you’re excited to write about it. The writing process is usually much faster when this happens and you’re writing “hot”. You may even be in a potential home run scenario because if you’re excited about something, your readers probably will be, too. You can always come back to the “old” idea. If it no longer seems worth completing, your revealed preference is probably telling you something.
What about if there’s an idea that’s a little off-topic? Say your newsletter is mostly movie reviews. But then you have a keen insight into parenting. Or you’re fascinated with a new book that you read. Or about the burgeoning rivalry between two chains of pizza restaurants in your town.
I’d probably go ahead and write it up, in the nose-to-tail spirit. As Max Read says:
What most successful Substacks offer to subscribers is less a series of discrete and self-supporting pieces of writing--or, for that matter, a specific and tightly delimited subject or concept--and more a particular attitude or perspective, a set of passions and interests, and even an ongoing process of “thinking through,” to which subscribers are invited.
Being in somebody’s inbox several times a week is an intimate thing. You’re letting people into your thought process. So, if you’re thinking, “I know I’m a movie reviewer, but this pizza story is really great,” I’d follow your instinct. (If I were subscribing to your newsletter, I’d want to read the pizza story.) This is one of the joys of being independent. If the New York Times’s movie reviewer wants to review pizza, they will have to jump through a million hoops to do it. I strongly advise maintaining a high quality threshold — especially when venturing off-topic6 — but otherwise, trust your process.
6. Writing well quickly is essential
Let’s say you came to me and are thinking about leaving your stuffy media job to start a Substack. Here’s the first thing I’d ask you: Can you reliably write clean copy relatively quickly, with little or no editing help?
If the answer is “yes,” you should probably do the newsletter. If not, you might use it as a bluff in your contract negotiations — but you probably shouldn’t.
Writing well quickly is a force multiplier for everything you do in your Substack. It’s Matt Yglesias’s big secret, for instance. Although you may encounter some diminishing returns after two to three posts per week, in general there’s no hard upper limit on the number of good posts that your audience will have an appetite for: we were publishing 5-ish times a week during the election peak, plus updating the model every day, and email open rates and the number of pageviews per story remained consistently high.
Just to underscore this, I am not saying that you should turn out some barely legible story in two hours if you could write a good story in four hours. But if you can turn around a good story in two hours, that’s nice.
The good news is that writing benefits from repetition and practice. During the pandemic, I went on runs five or six times a week. Running is neat because your time is pretty deterministic relative to your effort and fitness level that day. I’d regularly set new personal bests, going from not being able to run a 5K at all to something semi-respectable.7 Writing is like that, too. You’ll get much faster if you’re in the habit of writing regularly.
This is part of why I recommend the Always Be Blogging (ABB) approach — you’ll want to write enough so that it’s second nature to you. The other advantage of ABB is that you’ll already have done a lot of the prep work. So you’ll turn out 2000 pretty good words in two hours and wonder how you did that — and the answer is that you actually put in more than two hours because you spent a lot of time thinking about the thesis ahead of time. The process is the marathon: writing is the sprint that kicks in at the 25-mile mark.
7. Play the long game
One thing nearly every successful Substacker I’ve spoken with suggests is that their subscriber numbers were initially disappointing. Building up awareness of your new brand will probably take at least six months. At least. But the most successful Substacks continue to add customers for years.
Honesty goes a long way. If you’re in a potential home-run scenario — major news drops in one of your core topics — but you’re having trouble writing because you’re coming back from a bachelor party weekend in Las Vegas, you can actually just tell your readers that if you want to.
But beware of audience capture. To a large degree, you’re training your readers for what to expect. If you're not careful, this can become a self-fulfilling prophecy. If you intend for your two major “beats” to be the NBA and pizza reviews, and the first ten posts you write are pizza reviews, then you shouldn’t be surprised if the first NBA post you write doesn’t click well.
Earlier, I suggested looking at a variety of metrics to evaluate how well your posts did — even though your subscriber numbers are the ultimate bottom line. What they can miss, though, is that retaining readers is exceptionally important. Once your newsletter hits a certain level of maturity, keeping existing subscribers happy is the most important priority. In that spirit, thanks again so much for your support, and we hope that you’ll be with us for the long run.
I’m not exactly sure how Substack ranks its newsletters but I’m pretty certain that a calculation called “gross annualized revenues” gets a lot of weight. Basically, that’s the sale price of your total number of annual subscriptions plus your total number of monthly subscriptions, assuming monthly subscribers remain signed up for a year. It includes money that doesn’t flow to the creator — Substack’s 10 percent fee, plus Stripe (credit card processing) fees, which in principle are ~3 percent but in practice a bit higher given chargebacks and the like. Those fees apply to every newsletter, not just Silver Bulletin.
The difference in our case is that we’re much more weighted toward monthly subscriptions than most Substacks and — as much as we’d love to be proven wrong — we’re not expecting everybody who signed up on a monthly plan for our election coverage to remain subscribed. When people end their subscriptions, they have the opportunity to write a note explaining why. A lot of the notes are quite nice — along the lines of “really enjoyed your election coverage, see you in 2026/28!” — but convey that subscribers aren’t quite seeing enough value in the newsletter outside of elections. Which is … great actually! We basically have three product lines: essays/newsletters, election models/forecasts, and sports models/forecasts. The essays product line is highly evergreen, the election product line is highly cyclical, and sports is somewhere in between (sports are seasonal but played every year rather than every two/four years, plus we still need to build out our sports product). A great bakery that also serves great ice cream is going to see its ice cream business and therefore likely its overall revenues peak in the summer, but it still very much appreciates having customers every summer who are only there for the ice cream.
At least in terms of writing about politics; I’d written about baseball on the Internet for several years before then.
Sure, my politics have drifted apart from the Daily Kos set, but I swear it’s mostly them and not me! When I was blogging there in 2007/2008, the progressive ecosystem was more tolerant of heterodox political views and wonky analysis even when it didn’t support their conclusions.
That’s not to say you should deliberately seek out conflict — but if you write about politics and are any good at it, you’re inevitably going to get some. The same holds if you make honest election forecasts. In the long run, each party wins elections about half the time. But most consumers of political news are partisans and just want to hear good news. So if your forecasts are accurate, about half of the prospective audience will be annoyed with you at any given time.
The mailing list of unpaid subscribers is by far your best marketing channel: conversion usually goes browsing/lurking → free subscription → paid subscription.
A big exception here is if you don’t regularly write about politics and you’re thinking about a politics post. As someone who does write about politics, I know where the landmines are: that say, Gaza or trans rights or COVID are disproportionately likely to trigger flame wars in the comments section. I’ll sometimes proceed despite that, being aware of the hazards. (Not on the other two topics really but occasionally on COVID). If you don’t, it’s risky. I’ll sometimes see a sportswriter, for instance, tweet out something about politics and will cringe a little bit even when I agree with the sentiment because the author is unknowingly putting out a much hotter take than they were intending.
I’ve since regressed!
I think you have a fourth product too - you are really good at explaining modeling issues. That need not be limited to election models! (I teach a masters level quantitative methods class and I used lots of examples from the newsletter and encouraged my students to read/subscribe). There are loads of models being (mid)used out there in public policy - it would be great to have you talk to people in those areas and discuss the pros and cons of the models. You could start with the excellent Roger Pielke Jr, who has a great substack on climate issues and sports. I’d love to read a transcript of you two talking modeling!
I don’t care about sports. Is there a way to opt out of your sports newsletters? I love when you analyze anything else, especially politics. But no sports, thanks. 💜